RURAL-URBAN DIFFERENTIAL IN MATERNAL MORTALITY ESTIMATE IN NIGERIA, SUB-SAHARAN AFRICA

Abstract

In developing countries, the traditional sources of demographic statistics in which the estimates of demographic indices are based are either non-existence or incomplete. Data requirements on maternal deaths are always very large and costly. The indirect method (sisterhood method) for estimating maternal deaths was designed primarily as check to these problems. The study used Nigeria Demographic and Health Survey (NDHS), 2008 data. A total of 18,250 (6,894 urban and 11,356 rural) adults responded to questions essential for the estimation of maternal mortality. The P/F ratio method was used to adjust the total fertility rates (TFR) in urban and rural areas. Thereafter, the lifetime risks of maternal deaths (LTRMD) were estimated for the two areas. These were later converted to maternal mortality ratio (MMR). Data analyses revealed that the adjusted total fertility rates for urban and rural areas were 5.26 and 7.12 respectively. The LTRMD in urban was 0.0221 (1 in 45) whereas, in rural area it was 0.0309 (1 in 32). These results correspond to MMR of 424/100,000 and 440/100,000 live births in urban and rural areas respectfully. These are not far from the national estimate of 436/100,000 live births as evidence in this study. This method provided a robust estimate of MMR in both urban and rural areas and shows that the MMR in Nigeria is reducing. However, the figures at the two locations are still high. Government and international agencies should put appropriate mechanisms in place for further reduction in the prevalence.

Full text

74
RURAL-URBAN DIFFERENTIAL IN MATERNAL MORTALITY ESTIMATE IN NIGERIA,
SUB-SAHARAN AFRICA
Adebowale S. A; Fagbamigbe F. A; and Bamgboye E. A
Department of Epidemiology, Medical Statistics and Environmental Health
University of Ibadan, Ibadan, Nigeria
Email:adesteve2006@yahoo.com
ABSTRACT
In developing countries, the traditional sources of demographic statistics in which the
estimates of demographic indices are based are either non-existence or incomplete. Data
requirements on maternal deaths are always very large and costly. The indirect method
(sisterhood method) for estimating maternal deaths was designed primarily as check to these
problems. The study used Nigeria Demographic and Health Survey (NDHS), 2008 data. A
total of 18,250 (6,894 urban and 11,356 rural) adults responded to questions essential for
the estimation of maternal mortality. The P/F ratio method was used to adjust the total
fertility rates (TFR) in urban and rural areas. Thereafter, the life-time risks of maternal
deaths (LTRMD) were estimated for the two areas. These were later converted to maternal
mortality ratio (MMR). Data analyses revealed that the adjusted total fertility rates for urban
and rural areas were 5.26 and 7.12 respectively. The LTRMD in urban was 0.0221 (1 in 45)
whereas, in rural area it was 0.0309 (1 in 32). These results correspond to MMR of
424/100,000 and 440/100,000 live births in urban and rural areas respectfully. These are not
far from the national estimate of 436/100,000 live births as evidence in this study. This
method provided a robust estimate of MMR in both urban and rural areas and shows that the
MMR in Nigeria is reducing. However, the figures at the two locations are still high.
Government and international agencies should put appropriate mechanisms in place for
further reduction in the prevalence.
Keywords: life-time risks, adjusted total fertility rate, maternal mortality.
INTRODUCTION
Maternal mortality (MM) level is a part of indicators for assessing; overall health conditions,
reproductive health programs and development status in any Nation. However, few countries
have been able to establish a comprehensive reporting needed for its estimation. Maternal
mortality remains a major challenge to health systems worldwide. Reliable information about
its rates and trends is essential for resource mobilization and assessment of progress towards
Millennium Development Goal 5, the target for which is a 75% reduction in the maternal
mortality ratio (MMR) from 1990 to 2015.
The effort to lower maternal death rate in Nigeria has become a high government priority.
This informed the launching of the National Programme for the Prevention of Maternal
deaths. The aim was to expand and strengthen advocacy projects for safe motherhood.
Therefore, in order for maternal health programmes to remain focused, and to make a
quantitative evaluation of programme results, MM statistics are needed within segments of
the population.
Volume 2, September 2010
© 2010 Cenresin Publications
www.cenresin.org
Journal of Medical and Applied Biosciences

75
In most developing countries including Nigeria, the traditional sources of MM statistics (vital
registration system and sample surveys) in which the estimate of MMR is based are
unreliable and completely imperfect that the estimate obtained directly from such data are
often flawed and misleading. Also, results from hospital based studies are rarely acceptable
because the women who died in the facility are not representative of the population.
Therefore, the MMR obtained from such data is most likely to be biased. In such situations,
population-based surveys have to be used for its estimation.
In furtherance to these, it is understood that the best estimates of MM do not capture all
deaths related to pregnancy. However, there is strong evidence that official statistics
seriously under-estimate MM even in developed countries. While statistics on MMR are far
from perfect, they provide evidence of its magnitude around the world. Moreover, data
requirement on maternal deaths are always very large, which may involve 200,000
households and sometimes, follow up studies may be needed to track down the actual
number. Such data are always difficult to generate in developing countries in terms of cost
implication, time, and logistics and may be unrealistic in countries with small number of
inhabitants.
These problems compelled demographers to search for more efficient, cost effective and
refined means of measuring MM. One such method is the sisterhood method which was
originally designed to curtail the problem of large data requirements and cost. It is an
indirect method which based its analysis on four simple questions that asks adult
respondents about how many of their sisters have died and whether those who died were
pregnant at the time of death. The term indirect approach produces estimates of
demographic indices based on data or information that is indirectly related to its value. It is
the term used to describe estimation method that depends upon models or uses consistency
checks, or indeed uses conventional data in an unconventional way [1].
The original (indirect) sisterhood method was developed in the late 1980s (Graham, et-al,
1988) as an efficient means of measuring MM through population-based surveys, generating
a variety of indicators: the proportion of maternal deaths among female deaths, the MMR,
the maternal mortality rate and the lifetime risk of maternal death (LTRMD). The first field
trial of the method was carried out in September, 1987 in Gambia and has been used in
many studies. Therefore, its reliability and validity have been evaluated at different
international fora and conferences.
This current study adopted the sisterhood method to see rural-urban differential and national
estimate of MMR in Nigeria using adjusted total fertility rate (Adj.TFR). These estimates will
assist the planners and policy makers in their programs aimed at reducing MM in Nigeria. All
the techniques involved were strictly indirect.
Maternal Mortality in Nigeria
Most developing countries have no national statistics regarding MM and there are few studies
on MM in Nigeria. Although different studies on MM have been carried out in sub-section of
Adebowale S.A., Fagbamigbe F. A., and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

76
the country in the past, among which are the result of research from hospital records such as
[2][3][4] Most births in Nigeria do not take place in hospitals; therefore, the reported statistics
do not accurately reflect the numbers of deaths during pregnancy and childbirth. Hospital
statistics in Nigeria also suffer serious biases owing to selectivity and often lead to over or
under estimate of the level of MM.
Nigeria ranked second globally as the country with the highest estimated number of maternal
deaths with 37,000 cases of maternal deaths. The trend has shown an evidence of reduction.
For instance, in 2003 NDHS, it was estimated as 800/100 000 live births, whereas, in 2008
NDHS, the figure was 545/100 000 live births. Despite the reversal in the trend, the rate is
still considered to be high as indicated by WHO MM estimation guideline, 1997.
The high rate of MM in Nigeria is due to numerous causes which can be classified as either
direct or indirect obstetric. The direct obstetric causes are related to complications of
pregnancy, labor or in the 42-day post-partum period (puerperium), from incorrect treatment
or interventions e.g. haemorrhage, sepsis, eclampsia, obstructed labor, unsafe abortion. The
indirect obstetric causes are those resulting from a pre-existing disease or one that
developed during pregnancy and that is aggravated by pregnancy e.g. anemia, malaria,
cardiovascular disease, hepatitis, diabetes etc.
Studies have shown that flaws in the health care system and hostilities between midwives
and traditional birth attendants (TBAs) are contributing to the high rate of MM in Nigeria.
While many existing formal maternity services are underutilized, women in rural areas remain
under served. This reflects inaptness between the services being offered and the needs of
women. Among barriers to maternal care are the poor quality of services received at health
centers and attitudes of many health care providers. Health care providers are seen to be
unnecessarily harsh.
Nigerian women have two maternal health care systems available to them. The orthodox or
“modern” system of care is available in public and private maternity clinics, health centers,
special maternity hospitals and maternity units of general and specialist hospitals. The
traditional system comprises of healers and traditional birth attendants. Many women,
especially in the rural areas, patronize traditional birth attendants (TBAs). Even where
doctors are available, some women prefer TBAs because they are more familiar, accessible
and often less expensive than modern practitioners. In Nigeria, the cost of seeking modern
medical care is high and patients must often bring their own supplies to a hospital in order to
be treated. Facilities are often overcrowded, under-staffed and poorly equipped. In some
parts of Nigeria, women often prefer female health care providers and where they are not
readily available, they would rather stay away than allow men to treat them. In effect,
complicated cases are usually referred to hospital or health centers-often too late and many
women do die there.
Some cultural practices also make child-bearing risky and expose women to the danger of
death. There are other traditional beliefs and practice which contribute, less directly, to
Volume 2, September 2010
Journal of Medical and Applied Biosciences

77
increase risk of MM. These include nutritional taboos during pregnancy forbidden pregnant
women from eating some foods, early marriage, and early motherhood [5]. Many Nigerian
women live and work under cultural/religious conditions that do not allow them to reach their
full potential. They are not allowed to take decisions for their health needs or in their
reproductive lives, nor do they enjoy good health care.
In Nigeria, the TFR is high (5.7, [6]), this shows the transition to low fertility is yet to begun.
Nigerian society places great value on child bearing and parenthood and any couple which
fails to procreate becomes stigmatized and losses self-esteem. Therefore, women may suffer
from maternal depletion syndrome, whereby a woman’s health is compromised by numerous
and frequent pregnancies, food shortages and too much on child care and rearing.
Literature Review
The estimated number of maternal deaths for the world in 2000 was 529 000. These deaths
were almost equally divided between Africa (251 000) and Asia (253 000), with about 4%
(22 000) occurring in Latin America and the Caribbean and less than 1% (2 500) in the more
developed regions of the world. In terms of the MMR, the world figure is estimated to be 400
per 100 000 live births. By region, the MMR was highest in Africa (830), followed by Asia
(330), Oceania (240), Latin America and the Caribbean (190), and the developed countries
(20).
A comparable country, regional, and global estimates of MMR for 2005 was done to assess
trends between 1990 and 2005. The findings showed that there were 535 900 maternal
deaths in 2005, corresponding to a MMR of 402 (uncertainty bounds 216–654) deaths per
100000 live-births. Most maternal deaths in 2005 were concentrated in sub-Saharan Africa
(270 500, 50%) and Asia (240600, 45%). For all countries with data, there was a significant
decrease of 2·5% per year in the maternal mortality ratio between 1990 and 2005; however,
there was no evidence of a significant reduction in MMRs in sub-Saharan Africa in the same
period. The study also revealed that, some regions have shown some progress since 1990 in
reducing maternal deaths. Maternal mortality ratios in sub-Saharan Africa have remained
very high, with little evidence of improvement in the past [6].
Between 1980 and 2008 a database of 2651 observations of MM for 181 countries was
constructed using vital registration data, censuses, surveys, and verbal autopsy studies. They
used robust analytical methods to generate estimates of maternal deaths and the MMR for
each year between 1980 and 2008. The result of data analysis shows that there were
342900 maternal deaths worldwide in 2008, down from 526300 in 1980. The global MMR
decreased from 422 in 1980 to 320 in 1990, and was 251 per 100000 live-births in 2008.
The yearly rate of decline of the global MMR since 1990 was 1·3%. During 1990–2008, rates
of yearly decline in the MMR varied between countries, from 8·8% in the Maldives to an
increase of 5·5% in Zimbabwe. More than 50% of all maternal deaths were in only six
countries in 2008 (India, Nigeria, Pakistan, Afghanistan, Ethiopia, and the Democratic
Republic of the Congo) [7].
Adebowale S.A., Fagbamigbe F. A ., and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

78
The sisterhood method was applied to Djibouti population data. The survey was implemented
in February 1989. The results of the 7408 females 15-49 years interviewed shows that the
lifetime risk of dying of maternal causes were found to be 0.049 or 1 in 20.Using a total
fertility rate of 6.8, the MMR was calculated to be 740 maternal deaths per 100 000 live
births 11.6 years prior to the survey. The results of the assessment of the quality of the data
showed underreporting of the youngest age groups, which suggest misreporting. In spite of
the difficulties, the results are plausible and lend support to the method [8].
In Thyolo district in southern Malawi, 5 field teams used the sisterhood Method to interview
4124 people older than 15years in 7 traditional authorities to estimate the lifetime risk (LTR)
of maternal death and the MMR in this area. The life time risk of maternal death stood at 1 in
36 (1.0282).The MMR was 409/100 000 live births. These findings prove useful to community
and health leaders in designing intervention strategies to reduce MM in the area [9][10].
In another setting in Africa, the sisterhood method was used in a study carried out in rural
Niger. It involved 3058 respondents who identified 5796 sisters, among whom 186 were
reported to have died from maternal causes. Based on the study findings, the MMR was
estimated at between 1030 and 1050 per 100 000 live births, significantly higher than the
World Bank estimate of 700/100 000 live births for this part of Africa, but similar to rates
obtained using the same method in other West African countries with deficient data
collection.
The level of MMR estimated by the sisterhood method is presented for a rural district in the
Morogoro region of Southeastern Tanzania and the main causes of maternal death were
studied [11]. In this study, 4734 women in the Morogoro Region of Southeastern Tanzania
were interviewed using the sisterhood method to estimate MM. The resulting MMR of 448
deaths per 100 000 live births is much higher than the Tanzanian government's estimate for
the region, but much lower than the levels estimated by WHO and UNICEF. Advantages of its
use in this setting are that it is relatively cheap and feasible to obtain and is useful for small
areas where specific health information may not exist, such as the Kilombero valley.
Results of the sisterhood method have been proved to be fairly good when compared with
those derived through longitudinal surveys. For example, in Mwanza and Tanzania,
comparison of the MMR derived from a prospective community-based survey, the sisterhood
method survey, and hospital data, showed that the sisterhood method was fairly close to the
prospective community-based survey [12].
Studies on estimates of maternal mortality ratio carried out in two districts of the Brong-
Ahafo region, Ghana using sisterhood method were reviewed in 2000 [13]. Indirect estimates
of MMR were calculated from data collected in 1995 by family Health International (FHI) on
5202 women 15-49 years, using a household screen of randomly selected areas in the two
districts. Based on the Family Health International data, the MMR was estimated to be 269
maternal deaths per 100 000 live births for both districts combined [14]. The national MMR for
Ghana to be 214 MM per 100 000 live births, using indirect sisterhood data from a national
Volume 2, September 2010
Journal of Medical and Applied Biosciences

79
representative survey conducted in 1992 [15]. This figures was lower than a recent estimates
of Ghana’s MMR 742 maternal deaths per 100 000 live births.
Lech used sisterhood method to estimate maternal mortality in Swaziland by obtaining data
on 'sisterhood mortality from the 1993–1994 Multi-Purpose Household Survey carried out by
the Central Statistics Office and Ministry of Health of Swaziland [16]. A total fertility rate of
6.36, as given in the 1986 Swaziland census, was used in estimating these indicators. Prior to
this study, the maternal mortality rate (MMR) in Swaziland (based only on health facility
data) was considered to lie within the range of 107–125 maternal deaths per 100 000 live
births. The study revealed MMR to be 229 per 100 000 live births and the life-time risk of
maternal death to be 1 in 69.
In a study conducted in Shagamu, western part of Nigeria by Oladapo and others in 2006 to
investigate maternal deaths where all maternal deaths were recorded at Olabisi Onabanjo
University Teaching Hospital, Sagamu Nigeria in 2005 were retrospectively reviewed [17].
Information was obtained from a combination of admission and discharge registers, labour
and delivery records and retrieved case files from the Medical Records Department of the
hospital. The study revealed sixty-three (84.0%) of the deaths were direct maternal deaths
while 12 (16.0%) were indirect maternal deaths. Major causes of deaths were hypertensive
disorders in pregnancy (28.0%), haemorrhage (21.3%) and sepsis (20.0%). Overall,
eclampsia was the leading cause of deaths singly accounting for 24.0% of all maternal
deaths. Abortion and HIV-related mortality accounted for 1.3% and 4.0% of maternal
deaths, respectively. The research further showed that maternal mortality ratio of 2989.2 per
100 000 live births was significantly higher than that reported for 1988–1997 in the same
institution.
A ten-year review of maternal death in the University College Hospital, Ibadan Nigeria 1974,
showed that 820/100 000 maternal deaths occurred in the hospital during the period from
January 1, 1962, and December 31, 1971. However, the number of maternal deaths recorded
was not a true representative of what happens in the community since 60% of deliveries in
Nigeria take place outside the health facility [18].
Ujah [19] reviewed all the records of all deliveries and case files of all women who died during
pregnancy and childbirth between January 1, 1985 and December 31, 2001, in the maternity
unit of Jos University Teaching Hospital, Jos, Nigeria. The study showed a detailed and
comprehensive record-keeping of all deliveries, including complications and maternal deaths,
kept in the labour, antenatal, postnatal and caesarean section wards. A total of 267 maternal
deaths occurred among 36,768 deliveries over 17-year period, making the maternal ratio
(MMR) 740/100000 total deliveries. The trend fluctuates between 450 in 1960 and
1010/100000 deliveries in 1994.
In Nigeria [20], all maternal deaths were recorded at Ebonyi state university Teaching Hospital
(EBSUTH) Abakaliki, Nigeria; from January 2000 to December 2003. It was observed that
4192 live births were recorded, out of which 79 maternal deaths were obtained. It implies a
Adebowale S.A., Fagbamigbe F. A., and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

80
maternal mortality ratio of 1884 per 100 000 live births. This finding far exceeds the Nigerian
national average. The case records of only 49 (62%) of these maternal deaths were
complete and included in this review. This shows one of the inadequacies of hospital data in
estimating maternal mortality.
A population–based study was carried out to determine the incidence and causes of maternal
mortality as well as its temporal distribution over the last decade (1990-1999) in Kano [21].
This was a retrospective study using information contained in the vital statistics register
maintained by the research and statistics department of the Ministry of Health in Kano. The
village or local government council also reported births and deaths that occurred at home to
the Zonal council in charge of the area. All the maternal deaths recorded within the study
period in the Kano state, Nigeria, were analyzed. A total of 4154 maternal deaths occurred
among 171 621 deliveries, yielding a MMR of 2420 deaths per 100 000. Eclampsia, ruptured
and aneamia were responsible for about 50% of maternal deaths. The highest maternal
mortality ratio ever reported in the world was found.
MATERIAL AND METHODS
Sample Design
The data for this study is secondary and was obtained from ICF Macro Calverton, Maryland,
USA. It is an NDHS data, 2008. A brief description of the methodology involved during data
collection is discussed below.
The sample was designed to provide population and health indicators at the national, zonal,
and state levels. The primary sampling unit (PSU), referred to as a cluster for the 2008
NDHS, was defined on the basis of Enumeration Areas (EAs) from the 2006 EA census frame.
The 2008 NDHS sample was selected using a stratified two-stage cluster design consisting of
888 clusters, 286 in the urban and 602 in the rural areas. A representative sample of 36 800
households was selected, with a minimum target of 950 completed interviews per state. In
each state, the number of households was distributed proportionately among its urban and
rural areas.
All women age 15-49 and men age 15-59 who were either permanent residents of the
households in the 2008 NDHS sample or visitors present in the households on the night
before the survey were eligible to be interviewed. However, men were selected in a sub-
sample of half of the households. Three questionnaires were used. These are; the Household
Questionnaire, the Women’s Questionnaire, and the Men’s Questionnaire. These
questionnaires were adapted to reflect the population and health issues relevant to Nigeria.
Methods of Analysis
The method used in this study is multi-indirect which involves two procedures. First, the
adjusted total fertility rate (Adj.TFR) was estimated using Coale and Trussell P/F ratio model,
an indirect approach. The adjustment technique was based on questions on the total number
of women in each five-year age group, the number of children ever born and the number of
children born a year preceding the survey. The adjustment of the level of observed age-
Volume 2, September 2010
Journal of Medical and Applied Biosciences

81
specific fertility rates is necessary since they are assumed to represent the true age pattern
of fertility. This is done by combining the age pattern of the period fertility rates with the
level implied by the average parities of younger women to obtain a set of fertility rates that is
generally more reliable than either of its constituent parts.
Second, the life-time risk of maternal death was estimated using sisterhood method. The
estimation was possible through questions asked from adult respondents (men and women
aged 15 years and above) about the survival status of all their adult sisters born to the same
mother and whether these dead sisters were pregnant at the time of death. These data were
used to obtain the proportion of sisters dying during pregnancy, childbirth, or up to 6 weeks
after the end of pregnancy. Thereafter, standard adjustment factors were used to transform
those proportions into estimates of MM. The principal indicator obtained is the life-time risk
of maternal death which was converted to an estimate of the MMR by using the adjusted
total fertility rate. The formula is as shown below;
MMR = 1 (1 LTRMD)1 adj .TFR

where, LTRMD is the life time risk of maternal deaths.
RESULTS
Computational procedures for Adjusted Total Fertility Rate using P/F ratio:
Using this method the following steps were taken;
1. Average parities reported P(i):
Pi=Total number of children ever born to women in age group (i)
Total number of women in age group (i)
The denominator includes all women in age group i irrespective of the marital and
fertility status of the women.
e.g P(1) = 1527 6493 = 0.2352; P2=7310 6133 = 1.1919

2. Preliminary fertility schedule f(i):
fi=Number of births in the year preceding the survey in age group (i)
Total number of women in age group (i)
f1=630
6493 = 0.0970
f2=1540
6133 = 0.2511
·
·
·
3. Cumulated fertility schedule for a period 
Φi= 5 × f(j)
i
j=0 
Φ1= 5 × 0.0970 = 0.4850
Φ2=5×0.0970 + 0.2511= 1.7405
Φ3=5×0.0970 + 0.2511 + 0.2935= 3.2080
Adebowale S.A., Fagbamigbe F. A., and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

82
4. Average parity equivalents for a period (F(i)):
F(i) are computed by interpolation using the period fertility rates f(i) and the
cumulated fertility values Φi. Different techniques have been proposed for the
interpolation. Among the contributors to the methods are; Brass, Coale and Trussel.
While a simple polynomial model of fertility to know the relationship between
cumulated fertility schedule and average parity for successive age groups was fitted by
Brass. Coale and Trussel fitted a second degree polynomial which yielded equation (i)
below.
Fi=Φi1+ aifi+ bifi + 1+ ciΦ7 (1)
for i = 1,2,3, and ai, bi, ciare constants and are shown in APPENDIX I
F1= 0.0 + 2.5310.09700.1880.2511+ 0.00246.5655= 0.2741
F2= 0.4850 + 3.3210.25110.7540.2935+ 0.01616.5655= 1.2033
F3= 1.7405 + 3.2650.29350.6270.2868+ 0.01456.5655= 2.6142
F4= 3.2080 + 3.4420.28680.5630.2134+ 0.00296.5655= 4.0941
F5= 4.6420 + 3.5180.21340.7630.1187+ 0.00066.5655= 5.3061
F6= 5.7090 + 3.8620.11872.4810.0526+ 0.00016.5655= 6.0363
However, for F(7) the value is computed using;
F7=Φ6+ a7f7+ b7f6+ c7Φ7
F7= 6.3025 + 3.8280.05260.0160.1187+ 0.00026.5655= 6.5044
5. Fertility Schedule for conventional five-year age groups +():
f+(i) values are estimated by weighting the rates referring to unorthodox age groups using
the equation below:
f+i=1wi1fi+ wifi + 1 (2)
Where; wi= xi+ yi×fi
Φ7+ z(i) × f(i+1)
Φ7
The values of x(i), y(i) and z(i) are constants and are shown in APPENDIX II
NOTE: Childbearing is assumed to cease after age 50; there is no weighting factor i = 7
f+7=1w6f(7)
w1= 0.031 + 2.287 ×0.0970
6.5655 + 0.114 ×0.2511
6.5655 = 0.0692
w2= 0.068 + 0.999 ×0.2511
6.5655 0.233 ×0.2935
6.5655 = 0.0957
w3= 0.094 + 1.219 ×0.2935
6.5655 0.977 ×0.2868
6.5655 = 0.1058
w4= 0.120 + 1.139 ×0.2868
6.5655 1.531 ×0.2134
6.5655 = 0.1200
w5= 0.162 + 1.739 ×0.2134
6.5655 3.592 ×0.1187
6.5655 = 0.1535
w6= 0.270 + 3.454 ×0.1187
6.5655 21.497 ×0.0526
6.5655 = 0.1605
The values of w(i) are then substituted in equation (2) to give the following results;
f+1=10× 0.0970 + 0.0692 × 0.2511 = 0.1144
f+2=10.0692× 0.2511 + 0.0957 × 0.2935 = 0.2618
f+3=10.0957× 0.2935 + 0.1058 × 0.2868 = 0.2958
Volume 2, September 2010
Journal of Medical and Applied Biosciences

83
f+4=10.1058× 0.2868 + 0.1200 × 0.2134 = 0.2821
f+5=10.1200× 0.2134 + 0.1535 × 0.1187 = 0.2060
f+6=10.1535× 0.1187 + 0.1605 × 0.0526 = 0.1089
f+7=10.1605× 0.0526 = 0.0442
6. Adjustment of period fertility schedule:
This can be done by calculating the P/F ratios i.e average parity (column 5) divided by parity
equivalent (column 9). For example, for age group,
15-19
P(1)
F(1) =0.2352
0.2141 = 1.0986
20-24
P(2)
F(2) =1.1919
1.2033 = 0.9905
.
.
.
.
.
.
If the adjustment factor falls consistently between the age range 20-34, then the value of k
would be estimated as the average of P(2)/F(2), P(3)/F(3), and P(4)/F(4). i.e.
k = P(2) F(2)+
P(3) F(3)+
P(4) F(4)

3
Since, the adjustment factor does not fall consistently within this interval (i.e. 20-34). Then, k
can be computed as weighted average of P(2)/F(2) and P(3)/F(3). The weights are the
number of women in each age group as a proportion of women in both age groups
k = P(2) F(2)
×FP(2)
FP2+FP(3) + P(3) F(3)
×FP(3)
FP2+FP(3)
= 0.9905 ×6133
6133+6309 + 0.9619 ×6309
6133+6309 = 0.976
Then, the adjusted age-specific fertility rates for conventional age groups fi can be
estimated by simply multiplying the f+i values by the adjustment factor k. For example;
f1= kf+(1) = 0.976 × 0.1144 = 0.1117
f2= kf+(2) = 0.976 × 0.2618 = 0.2555
7. The adjusted total fertility rate for the total sample is then estimated as multiplying
the sum of age-specific fertility rate fi by 5
TFR = 5 × fi= 5 × 1.281 = 6.41
The same procedures were used for the computations of adjusted TFR for both rural and
urban areas. The values are shown in Tables 2 and 3.
Adebowale S.A., Fagbamigbe F. A and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

84
TABLE 1. REPORTED PERIOD AND ADJUSTED FERTILITY RATES FOR CONVENTIONAL AGE GROUPS,
NIGERIA, NDHS, 2008
1
2
3
4
5
6
7
8
9
10
11
Age
group
CEB(i)
Births
1-year
FP(i)
P(i)
f(i)
Φ(i)
F(i)
+()
P/F
()
15-19
1527
630
6493
0.2352
0.0970
0.4850
0.2141
0.1144
1.0986
0.1117
20-24
7310
1540
6133
1.1919
0.2511
1.7405
1.2033
0.2618
0.9905
0.2555
25-29
15864
1852
6309
2.5145
0.2935
3.2080
2.6142
0.2958
0.9619
0.2887
30-34
18256
1329
4634
3.9396
0.2868
4.6420
4.0941
0.2821
0.9623
0.2753
35-39
20578
835
3912
5.2602
0.2134
5.7090
5.3061
0.2060
0.9913
0.2011
40-44
18727
360
3032
6.1765
0.1187
6.3025
6.0363
0.1089
1.0232
0.1062
45-49
19651
151
2872
6.8423
0.0526
6.5655
6.5044
0.0442
1.0519
0.0431
TOTAL
33,385
1.3131
1.3132
1.2816
Total fertility rate…………
6.41
TABLE 2. REPORTED PERIOD AND ADJUSTED FERTILITY RATES FOR CONVENTIONAL AGE GROUPS,
URBAN NIGERIA, NDHS, 2008
1
2
3
4
5
6
7
8
9
10
11
Age
group
CEB(i)
Births
1-year
FP(i)
P(i)
f(i)
Φ(i)
F(i)
+()
P/F
()
15-19
250
118
2268
0.1102
0.0520
0.2600
0.1086
0.0629
1.0147
0.0621
20-24
1791
430
2261
0.7921
0.1902
1.2110
0.7672
0.2048
1.0325
0.2023
25-29
4679
678
2432
1.9239
0.2788
2.6050
2.0338
0.2821
0.9460
0.2786
30-34
5436
449
1709
3.1808
0.2627
3.9185
3.4282
0.2556
0.9278
0.2525
35-39
5986
232
1354
4.4210
0.1713
4.7750
4.4702
0.1616
0.9890
0.1596
40-44
5533
73
1028
5.3823
0.0710
5.1300
4.9502
0.0651
1.0873
0.0643
45-49
5284
35
882
5.9909
0.0397
5.3285
5.2820
0.0335
1.1342
0.0331
TOTAL
11934
1.0657
1.0656
1.0525
Total fertility rate…………
5.26
Volume 2, September 2010
Journal of Medical and Applied Biosciences

85
TABLE 3. REPORTED PERIOD AND ADJUSTED FERTILITY RATES FOR CONVENTIONAL AGE GROUPS,
RURAL NIGERIA, NDHS, 2008
1
2
3
4
5
6
7
8
9
10
11
Age
group
CEB(i)
Births
1-year
FP(i)
P(i)
f(i)
Φ(i)
F(i)
+()
P/F
()
15-19
1273
512
4225
0.3013
0.1212
0.6060
0.2702
0.1424
1.1151
0.1400
20-24
5525
1110
3872
1.4269
0.2867
2.0395
1.4465
0.2952
0.9865
0.2902
25-29
11179
1174
3877
2.8834
0.3028
3.5535
2.9445
0.3046
0.9793
0.2994
30-34
12835
880
2925
4.3880
0.3009
5.0580
4.4772
0.2972
0.9801
0.2921
35-39
14591
603
2558
5.7041
0.2362
6.2390
5.7840
0.2296
0.9862
0.2257
40-44
13174
287
2004
6.5739
0.1432
6.9550
6.6479
0.1317
0.9889
0.1294
45-49
14444
115
1990
7.2583
0.0578
7.2440
7.1771
0.0482
1.0113
0.0474
TOTAL
21451
1.4488
1.4489
1.4242
Total fertility rate…………
7.12
Computation of Maternal Mortality Ratio
The data used for the analyses of MMR in NDHS 2008 survey are; How many sisters have
you ever had, born to the same mother, who ever reached the age 15 (or who were ever
married) including those who are now dead? How many of these sisters reaching age 15 are
alive now? How many of these sisters are dead? How many of these dead sisters died during
pregnancy or during childbirth, or during the six weeks after the end of the pregnancy?
These questions are used to derive the proportions of adult sisters dying during pregnancy,
childbirth or puerperium. Standard adjustment factors were used to convert these
proportions into LTRMD which was later converted to MMR.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
15-19
20-24
25-29
30-34
35-39
40-44
45-49
Age specific fertility rate
Age
Figure. 1: Adjusted Age Specific Fertility Rates (f*(i)) for
Urban, Rural and Total Population in Nigeria, NDHS, 2008
Urban
Rural
Total
Adebowale S.A., Fagbamigbe F. A., and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

86
TABLE 4. ESTIMATION OF LIFE-TIME RISK OF MATERNAL DEATHS FOR CONVENTIONAL
AGE GROUPS, FOR NIGERIA, NDHS, 2008
1
2
3
4
5
6
7
Age
group
Number of
Respondents
Number
of sisters
15 years
and above
Maternal
Deaths
Adjustment
Factor
Sister
Units of
Risk
Exposure
Life-Time
Risk of
Maternal
Deaths
15-19
3,233
17,652*
68
0.107
1889
0.036
20-24
3,414
18,640*
157
0.206
3840
0.041
25-29
3,478
18,480
196
0.343
6339
0.031
30-34
2,684
14,912
191
0.503
7501
0.026
35-39
2,232
12,511
211
0.664
8307
0.025
40-44
1,684
9,362
184
0.802
7508
0.025
45-49
1,525
7,996
168
0.900
7196
0.023
TOTAL
1,175
42,580
0.0276
*Derived by multiplying the number of respondents by the average number of
ever-married sisters per respondent reported for the age groups 25+ i.e. 5.46.
(Reported numbers: 15-19 = 12063, 20-24 = 15591)

=
18480
3478
+
14912
2684
+
12511
2232
+
9362
1684
+
7996
1525
5
=
5
.
46
Maternal Mortality Ratio =11Lifetime risk of maternal deaths1
adj.TFR
=110.02761
6.41 =436 per 100 000
TABLE 5. ESTIMATION OF LIFE-TIME RISK OF MATERNAL DEATHS FOR
CONVENTIONAL AGE GROUPS, FOR URBAN NIGERIA, NDHS, 2008
1
2
3
4
5
6
7
Age
group
Number of
Respondents
Number of
sisters 15
years and
above
Maternal
Deaths
Adjustment
Factor
Sister
Units of
Risk
Exposure
Life-Time
Risk of
Maternal
Deaths
15-19
1187
6647*
20
0.107
711
0.028
20-24
1358
7605*
48
0.206
1567
0.031
25-29
1417
7518
64
0.343
2579
0.025
30-34
1060
5842
66
0.503
2939
0.023
35-39
807
4678
68
0.664
3106
0.022
40-44
595
3352
46
0.802
2688
0.017
45-49
470
2703
42
0.900
2433
0.017
TOTAL
354
16023
0.0221
*Derived by multiplying the number of respondents by the average number
of ever-married sisters per respondent reported for the age groups 25+ i.e.
5.60. (Reported numbers: 15-19 = 3862, 20-24 = 6101)
Adebowale S.A., Fagbamigbe F. A., and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

87
Maternal Mortality Ratio =11Lifetime risk of maternal deaths1
adj.TFR
=110.02211
5.26 =424 per 100 000
TABLE 6. ESTIMATION OF LIFE-TIME RISK OF MATERNAL DEATHS FOR CONVENTIONAL AGE
GROUPS, FOR RURAL NIGERIA, NDHS, 2008
1
2
3
4
5
6
7
Age
group
Number of
Respondents
Number of
sisters 15
years and
above
Maternal
Deaths
Adjustment
Factor
Sister
Units of
Risk
Exposure
Life-Time
Risk of
maternal
deaths
15-19
2046
11028*
48
0.107
1180
0.0407
20-24
2056
11082*
109
0.206
2283
0.0477
25-29
2061
10962
132
0.343
3760
0.0351
30-34
1624
9070
125
0.503
4562
0.0274
35-39
1425
7833
143
0.664
5201
0.0275
40-44
1089
6010
138
0.802
4820
0.0286
45-49
1055
5293
126
0.900
4764
0.0265
TOTAL
821
26570
0.0309
*Derived by multiplying the number of respondents by the average number
of ever-married sisters per respondent reported for the age groups 25+ i.e
5.39. (Reported numbers: 15-19 = 8201, 20-24 = 9490)
Maternal Mortality Ratio =11Lifetime risk of maternal deaths1
adj.TFR
=110.03091
7.121 =440 per 100 000
DISCUSSION
Sisterhood method was carried out for the first time in the Gambia (1987). The result
indicates a lifetime risk of MM of 0.0584 or 1 in 17 which yielded a MMR of 1005 per 100 000
live births. This figure was higher than the previous estimate of MM in Gambia and the
method provides an approach suited for estimating MM at a national and sub-national level.
Sisterhood method of determining MM often provides data that are more comprehensive than
facility-based records.
In Nigeria MM is known to be high, yet a major problem is unavailability of sufficient data to
closely monitor the effectiveness of various interventions program [22]. This is because vital
registration system (VRS) in Nigeria is poor, thus affecting the availability of data on MM. The
VRS in Nigeria is poor because of its low level of completeness, reliability and validity.
Unfortunately, only three national surveys had addressed the issue of MMR in Nigeria; the
1999 and 2008 NDHS and the multiple cluster indicator survey. This has posed numerous
constraints on the effective; planning, management, monitoring and evaluation of maternal
Volume 2, September 2010
Journal of Medical and Applied Biosciences

88
mortality reduction strategies. This paper therefore, used a sisterhood method which has one
possible means of gauging the rural-urban MM level in Nigeria.
In developing countries, two methods are generally in use for the estimation of MMR. These
are; the direct and the indirect [22], otherwise known as sisterhood method. In the NDHS
2008 report, direct method which was based on the report from reported survivorship of
sisters for the six-year period before the survey was used to estimate MMR. Using the
appropriate procedures, the MMR was 545/100 000 live births. However, due to deficiencies
in the quality of data collection and reporting in developing countries like Nigeria, the present
study was carried out. The data underwent series of demographic adjustments for the
computation of TFRs that were used for the estimates. Moreover, the NDHS 2008 failed to
address the urban-rural differential in MMR. This study therefore, adjusted the TFRs for both
rural, urban and the total sample using the P/F ratio method. The results showed that the
Adj.TFR for rural, urban and total were 7.12, 5.26 and 6.41 respectively. These were used for
the estimation of MMR.
The LTRMD was 0.0309 (1 in 32) in rural and 0.0221 (1 in 45) in urban area. The estimated
MMR displayed a plausible pattern, being higher in rural (440/100 000 live births) than urban
(424/100 000) area. The national estimate of LTRMD and MMR were 0.0276 (1 in 36) and
436/100 000 live births respectively. If these figures are compared with international
specification for high and low risks of maternal deaths that a lifetime risk of 1 in 3000
represents a low risk of dying from pregnancy and childbirth, while 1 in 100 is a high risk13.
Therefore, the finding that for every 36 women, one will die of pregnancy and childbirth
related conditions in Nigeria is highly frightening.
The MMR in Nigeria as shown in this study appears to be at pal with estimates obtained in
other African countries. For instance, in year 2000, the estimate of MMR in Southeastern
Tanzania, was 448 per 100 000 live births [23] while the estimates for the two samples of the
population in rural Northern Tanzania were 362 and 444 per 100 000 live births respectively
[24]. Also, the estimate of MMR carried out in two districts of the Brong-Ahafo region of
Ghana in year 2000 was 269 maternal deaths per 100 000 live births for both districts
combined (WHO 2001).
In conclusion, there is slight urban rural differential in MMR in Nigeria. However, the figures
are still high at the two locations. Maternal mortality in Nigeria is reducing. Government and
international agencies should put appropriate mechanisms in place for further reduction in
the prevalence. The sisterhood technique is a simple and robust way of estimating MMR.
The methodology was based on assumptions designed several years ago which may not
really be applicable to the present time. Hence, the method should be refined to match on
with the present day demographic system.
Adebowale S.A., Fagbamigbe F. A., and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

89
ACKNOWLEDGEMENTS
The authors are grateful to National Population Commission who gave permission for the use
of NDHS 2008 data. We would like to thank Professor Olushola Ayeni for his contributions
and technical advice in the course of the paper write-up and those who made an input at one
time or the other.
REFERENCES
1 United Nations Publication. Indirect method of demographic estimation. Manual X 1984.
2 Ogunniyi S O, Faleyimu B L. “Trends in maternal deaths in Ilesa, Nigeria, 1977–1988” West
African Journal of Medicine. 1991;10(1):400-404.
3Okaro J, Umezulike A, Onah E, Chukwuali I, Ezugwu F, Nweke C. “Maternal mortality at the
University of Nigeria Teaching Hospital, Enugu, African Journal of Reproductive Health.
2001;5(2):90-97.
4 Etuk S J, Itam I H, Asuquo E E J. “Morbidity and mortality in booked women who
deliver outside orthodox health facilities in Calabar, Nigeria” Acta Tropica.
2000;75(3):309-313.
5 Federal ministry of health, Nigeria, 2005.
6 National Population Commission of Nigeria, Population and Housing Census Strategy and
Implementation Plan. Abuja, Nigeria: NPC; 2008
.
7Hill K. et al. Suzuki Estimates of maternal mortality worldwide between 1990 and 2005:
an assessment of available data. The Lancet, Volume 370, Issue 9595, Pg 1311-1319.
8Hogan M. Maternal mortality for 181 countries, 1980–2008: a systematic analysis of
progress towards Millennium Development Goal 5.
The Lancet
, 2009. Volume 375,
Issue 9726, Pages 1609-1623
9Songane F, Bergstrom S. Quality of registration of maternal deaths in Mozambique: a
community-based study. Social Science & Medicine 2002 Vol.
54: Pg23–31
10 Mswia R, Community-based monitoring of safe motherhood in the United Republic of
Tanzania. Bulletin of the World Health Organization
2003. 81(2): 87–94.
11Font F, et al. Maternal mortality in a rural district of southeastern Tanzania: an application
of the sisterhood method. International Journal of Epidemiology 2000. Vol.29: Pg 107–
112.
12Walraven Gel, Mkanje, R.J.B., Van Roosmalen J .,Van Dongen, P.W and Dolmans W.
Assessment of Maternal Mortality in Tanzania. Br.Journal of Obstetrics Gynaecology
1994 Vol 101: Pg 414-417.
Volume 2, September 2010
Journal of Medical and Applied Biosciences

90
13Smith J B, Fortney J A., Wong E, Amatya R, Coleman N A and de Graft Johnson J. Estimates
of the maternal mortality ratio in two districts of the Brong-Ahafo2000.
14World Health Organization, author. Maternal mortality in 2000: estimates developed by
WHO, UNICEF, UNFPA, World Health Organization, Geneva. 2004.
17Oladapo O T, Lamina A M and Fakoya T U. Maternal deaths in Shagamu in the new
millennium: A facility based retrospective analysis 2006.
18National Population Commission (NPC), and ORC Macro Nigeria Demographic and Health
Survey 2003. Calverton, Maryland: 2004. pg. 3
19Ujah et. al. Factors Contributing to Maternal Mortality In North –Central Nigeria. A
Seventeen year Review. African Journal of Reproductive Health 2005.
20Umeora O U, Esike C O, Egwuatu V E. Maternal Mortality in rural Nigeria. International
Journal of Gynaecology and Obtetrics 2005; Vol 88: Pg 321-322.
21Adamu, Y.M., Salihu H.M., Sathiakumar N. and Alenxander G.R Maternal Mortality in
Northern Nigeria: a population –based study. European Journal of Obstetrics,
Gynecology and Reproductive Biology 2003; 109(2):153-159.
22Rutenberg N, Sullivan J M, Survey D H. Direct and Indirect estimates of maternal mortality
from the sisterhood method .in demographic and Health Survey World Conference.
August 5-7, 1991.
23Font F, Alonso G M, Nathan R, Lwilla, Kimario J and Tanner M. Maternal mortality in a rural
district of southeastern Tanzania: an application of the sisterhood method.
International Journal of Epidemiology 2000; Vol.29: Pg 107–112.
24Olsen B E, Hinderaker S G and Kazaura M, Estimates of Maternal Mortality by the
Sisterhood Method in rural Northern Tanzania: a household sample and an antenatal
clinic sample. British Journal of Obstetrics and Gynaecology 2000; Vol. 107(10):1290- 7.
Adebowale S.A., Fagbamigbe F. A., and Bamgboye E. A.
Rural-Urban Differential in Maternal Mortality
Estimate in Nigeria; Sub-Saharan Africa

91
APPENDIX
APPENDIX II: COEFFICIENTS FOR CALCULATION OF WEIGHTING
FACTORS TO ESTIMATE AGE-SPECIFIC FERTILITY RATES FOR
CONVENTIONAL AGE GROUPS FROM AGE GROUPS SHIFTED BY SIX
MONTHS
Age Group
Index (i)
Coefficients
x(
i
)
y(
i
)
z(
i
)
15-19……………
1
0.031
2.287
0.114
20-24……………
2
0.068
0.999
-0.233
25-29……………
3
0.094
1.219
-0.977
30-34……………
4
0.120
1.139
-1.531
35-39……………
5
0.162
1.739
-3.592
40-
44…………....
6
0.270
3.454
-21.497
Adapted from United Nations Publication, Manual X
*This coefficient should be applied to
f
(
i
-1), not
f
(
i
+1), that
is, to
f
(6) instead of
f
(8)
APPENDIX I: COEFFICIENTS FOR INTERPOLATION BETWEEN
CUMULATED FERTILITY RATES TO ESTIMATE PARITY EQUIVALENTS
Age Group
Index (i)
Coefficients
a(i)
b(i)
c(i)
15-19……………
1
2.531
-0.188
0.0024
20-24……………
2
3.321
-0.754
0.0161
25-29……………
3
3.265
-0.627
0.0145
30-34……………
4
3.442
-0.563
0.0029
35-39……………
5
3.518
-0.763
0.0006
40-44…………....
6
3.862
-2.481
-0.0001
45-49……………
7
3.828
0.016*
-0.0002
Adapted from United Nations Publication, Manual X
*This coefficient should be applied to
f
(
i
-1), not
f
(
i
+1), that
is, to
f
(6) instead of
f
(8)
Volume 2, September 2010
Journal of Medical and Applied Biosciences